Non-traditional alternative sources of electric power may be configured as cells, such as photovoltaic or fuel cells, to produce dc current across a potential difference. FIG. 1 diagrammatically represents a photovoltaic array 100 comprising a plurality of electrically connected photovoltaic modules 102, each one of which comprises a plurality of electrically interconnected photovoltaic cells. Although a three column by “n” row (where n can be any positive integer) is depicted in FIG. 1, the photovoltaic array may comprise any number of rows and columns of photovoltaic modules. For convenience a photovoltaic array is referred to as a “PVA.” The PVA is a source of dc electric power (at output terminals +DC and −DC in FIG. 1) with performance characteristics graphically illustrated in FIG. 2. The voltage and current capacity of a PVA is a function of incident sunlight on the photovoltaic cells and their ambient temperature (site parameters). Additionally the level of current demand from the output of the PVA directly affects the magnitude of the dc output voltage. Curves 120a through 120e each represent change in PVA output current relative to a change in PVA output voltage for a particular type of photovoltaic cell and/or site parameters; curves 120a′ through 120e′ represent corresponding change in PVA output power relative to a change in PVA output voltage. As illustrated by the series of curves 120a through 120e, the current output gradually decreases as output voltage increases, until the maximum capacity of the PVA is reached for a given level of incident sunlight and ambient temperature. At that point in each current curve, the current output rapidly drops. Corresponding PVA power output is represented by the series of curves 120a′ through 120e′; the PVA output power increases with voltage output to a point defined as the “maximum power point” (MPP), as defined by the intersection of dashed line MPP and each power curve in FIG. 2, and then rapidly drops. Therefore the desired optimum operating point for a power generating PVA is the MPP point.
Generally then, a PVA represents a dc source having a degree of unpredictability in terms of output stability since the output is a function of instantaneously uncontrollable factors such as the instantaneous level of incident sunlight or ambient temperature.
To deliver electric power in appreciable quantities from a PVA to a traditional electric power system (referred to as a “grid”), PVA dc output power must be converted to ac power at the grid frequency and phase synchronized with grid power. Alternative “power plants,” such as a photovoltaic (solar) farm formed from a collection of photovoltaic arrays and dc-to-ac power converters, can have an electrical output capacity ranging from a few kilowatts to hundreds of megawatts. Solar farms are preferably built in regions with abundant sunlight, such as mountainous regions and deserts. Solar farms can also be built on roofs of high capacity power consumers, such as refrigerated storage facilities, industrial manufacturing plants, buildings housing banks of computer network servers, and shopping malls.
FIG. 3 shows a typical prior art, three phase, switch mode voltage source inverter 130. The inverter comprises three branches, each with two switching devices (SW1 and SW2; SW3 and SW4; or SW5 and SW6) in each branch. The switching device utilized in the switch mode voltage source inverter may be any type of controllable, unidirectional conduction semiconductor device, for example, a bipolar junction transistor (BJT); a metal-oxide-semiconductor field-effect transistor (MOSFET); an insulated-gate bipolar transistor (IGBT); a gate turn-off thyristor (GTO); or a gate commutated thyristor (GCT). Each switching device is shunted with an anti-parallel diode (D1 through D6). DC voltage input to the switch mode voltage source inverter is from PVA 100. Smoothing capacitor Cdc stabilizes the input dc voltage while the dc current instantaneously changes through each half cycle of the inverter output frequency. The switching devices are modulated by sequentially switching them from the conduction (on) to non-conduction (off) states at a high rate of several kilohertz, so that the inverter output current, after passing through ac low pass filter 132, will be close to an ideal sinusoidal waveform. The inverter output current is then transformed through line transformer 134, which electrically isolates the inverter output from grid 92 and transforms the inverter output voltage level to the grid voltage level. Current supplied to the grid feeds load Rload and, therefore, reduces the burden on grid power sources Vac, which supply power through grid impedance Zline.
The graphs in FIG. 4(a) through FIG. 4(c) describe the operation of the switch mode voltage source inverter shown in FIG. 3. A high frequency (typically 1,000 to 5,000 Hertz) saw tooth control signal represented by waveform 140 in FIG. 4(a) is compared with a sinusoidal reference signal represented by waveform 142, which is synchronized with the inverter output phase voltages. The waveforms illustrate one phase, for example phase A, while the other two phases, namely phases B and C, are identical except for a phase shift of plus and minus 120 degrees.
When the saw tooth signal's instantaneous magnitude is greater than the magnitude of the reference sinusoidal signal, positive switching device SW1 (connected to +DC rail) is conducting and negative switching device SW2 (connected to −DC rail) is non-conducting. At this instant a positive potential is applied to the inverter output phase. When the saw tooth signal's instantaneous magnitude is less than the magnitude of the sinusoidal reference signal, positive switching device SW1 is non-conducting and negative switching device SW2 is conducting. At this instant a negative potential is applied to the inverter output phase. Thus a high frequency pulse width modulated (PWM) train of voltage pulses represented by waveform 144 in FIG. 4(b) are generated on each inverter output phase.
For proper operation, the dc output voltage of the PVA must be at least equal to, or greater than, any peak value of the grid's phase voltages (Van, Vbn and Vcn) that are induced by the grid via transformer 134 onto inverter output phases A, B and C. To satisfy this condition, the amplitude of the generated saw tooth signal is equal to the amplitude of the generated reference sinusoidal signal. When the PVA dc output voltage is greater than the phase voltages, the amplitude of the reference sinusoidal signal is reduced below the peak saw tooth voltage and the PWM voltage is changed, thereby controlling the magnitude of the output current represented by waveform 146 in FIG. 4(c). When the PVA dc output voltage drops below peak phase voltages, the inverter controls cannot compensate for the low level dc, and the total harmonic content of the inverter output current becomes so great that inverter 130 is disconnected from the grid.
Switching devices SW2 and SW3 are controlled in a similar way, except that the sine wave control signal is shifted plus 120 degrees from that for phase A, so that the inverter output produces a plus 120 degrees-shifted phase B, PWM voltage and sinusoidal current; similarly for switching devices SW4 and SW5, the sine wave control signal is shifted minus 120 degrees from that for phase A, so that the inverter output produces a minus 120 degrees-shifted phase C, PWM voltage and sinusoidal current.
In the switch mode voltage source inverter, switching devices SW1 through SW6 are the only control elements responsible for the value and shape of the inverter output current supplied to the grid. They are switched (commutated) at a high rate, which requires high speed semiconductor devices that are limited in steady state current carrying capacity and power dissipation. Switching losses are the limiting factor for the amount of power that can be converted by this type of inverter. Although switch mode voltage source inverters are widely used in residential and some commercial solar power converters that are capable of generating power levels up to 500 kW, they are too small to be successfully used in multi-megawatt ranges needed for large solar farm customers. The main reason for maximum power limitation is the high frequency of commutation of switching devices SW1 through SW6 that contribute to significant power losses in the devices with limited capacity to dissipate these losses.
The above description of a three phase switch mode voltage source inverter may be implemented with various switching schemes that are based on a rigid dc voltage input and high frequency PWM commutation of the inverter switching devices.
It is one object of the present invention to convert dc electric power from a plurality of typically unstable dc sources with multiphase regulated current source inverters having multiphase transformed outputs that produce an ac output current with reduced total harmonic distortion for injection into an electric power grid.